Workshop: Bayesian Latent Class Models to evaluate diagnostic tests
This 3-day training will focus on the application of Bayesian Latent Class analysis Models (BLCMs) for field evaluation of diagnostic tests. Participants will learn about the theory of Bayes’ theorem, probability distributions, prior distributions, and introduction to simulation and Markov-chain Monte Carlo methods for simulating priors and posteriors, and how they can be applied for estimation of diagnostic test sensitivity and specificity when tests are imperfect, and a reference test is not available.
Outline and registration
The training is interactive through discussion and practical examples among participants and with instructors. English will be the working language of this training. Participants should have basic knowledge and experience in writing computing programs in the statistical language R. Participants will need to bring their laptops.
By attending this training, participants will:
- Perceive the logic of latent class models and their applicability in diagnostic accuracy studies in veterinary medicine
- Get acquainted with BLCMs basic principles & challenges
- Perform hands-on training on Se/Sp estimation with BLCMs
- Understand the importance of standards for reporting of diagnostic accuracy studies that use BLCMs (STARD-BLCMs)
Location
Lecture room Y17-M-05, Winterthurerstr. 190 8057 Zurich, Switzerland
Dates
14-16 July 2021
How to Register
https://forms.gle/3mqFYLLUBCedAVp3A
The first eligible 15 participants will receive an official invitation from e-COST.